Research Article

Unbiased yeast screens identify cellular pathwaysaffected in Niemann–Pick disease type CAlexandria Colaco1 , Marıa E Fernandez-Suarez1 , Dawn Shepherd1, Lihi Gal2, Chen Bibi2 , Silvia Chuartzman2,Alan Diot3 , Karl Morten3, Emily Eden4 , Forbes D Porter5, Joanna Poulton3, Nick Platt1 , Maya Schuldiner2 ,Frances M Platt1

Niemann–Pick disease type C (NPC) is a rare lysosomal storagedisease caused by mutations in either the NPC1 or NPC2 genes.Mutations in the NPC1 gene lead to the majority of clinical cases(95%); however, the function of NPC1 remains unknown. To gainfurther insights into the biology of NPC1, we took advantage ofthe homology between the human NPC1 protein and its yeastorthologue, Niemann–Pick C–related protein 1 (Ncr1). We recre-ated the NCR1 mutant in yeast and performed screens to identifycompensatory or redundant pathways that may be involved inNPC pathology, as well as proteins that were mislocalized inNCR1-deficient yeast. We also identified binding partners of theyeast Ncr1 orthologue. These screens identified several processesand pathways that may contribute to NPC pathogenesis. Theseincluded alterations in mitochondrial function, cytoskeletonorganization, metal ion homeostasis, lipid trafficking, calciumsignalling, and nutrient sensing. The mitochondrial and cyto-skeletal abnormalities were validated in patient cells carryingmutations in NPC1, confirming their dysfunction in NPC disease.

DOI 10.26508/lsa.201800253 | Received 23 November 2018 | Revised 21 May2020 | Accepted 22 May 2020 | Published online 2 June 2020

Introduction

Niemann–Pick disease type C (NPC) is an autosomal recessive ly-sosomal storage disorder characterized by progressive neuro-degeneration. NPC is caused by mutations in either the NPC1 orNPC2 genes, resulting in identical clinical phenotypes irrespectiveof which gene is affected (1). Mutations in NPC1 account for themajority of observed clinical cases (95%); however, the exactfunction of this protein remains incompletely understood. Thereare currently two main theories about NPC1 function; one is thatNPC1 is a cholesterol transport protein that moves low-densitylipoprotein-derived cholesterol out of the lysosome (2), whereasthe second is that NPC1 is a cholesterol-regulated protein that is

directly or indirectly involved in the transport of other lipid cargoswithin or across the lysosomal membrane (3). Structurally, NPC1 is a13 transmembrane domain protein that contains a sterol-sensingdomain and has structural similarities with resistance-nodulation-division permeases (multi-substrate effluxors) (4, 5). The highlyconserved structure of the NPC1 protein makes it a good target forstudies in simpler model eukaryotes that may provide novel in-sights into its conserved functions.

In the yeast Saccharomyces cerevisiae (here on referred to asyeast), the NPC1 orthologue is the Niemann–Pick type C–relatedprotein (Ncr1), which localizes to the vacuole, the yeast equivalentof the mammalian lysosome (6). Studies have demonstrated thatthe human NPC1 and yeast Ncr1 protein are functionally equivalent,as the cellular phenotypes of patient-derived fibroblasts can berescued through the overexpression of tagged yeast Ncr1 proteinthat directs it to the lysosomal membrane (6). It had previouslybeen reported that there is no significant change in sterol orphospholipid levels in NCR1 mutants (Δncr1), but rather a sphin-golipid trafficking defect (6) where long-chain sphingoid bases (7)accumulate in Δncr1 yeast. Further studies demonstrated that whilesterols may not accumulate in the vacuole in Δncr1 yeast (6), understarvation conditions, the processing of lipid droplets and transportof sterols to the vacuolar membrane is impaired (8). These data,implicating defects in sphingolipid and sterol trafficking, are in linewith the recent structural data identifying an internal hydrophobictunnel environment in Ncr1 that would accommodate a variety oflipids, in a capture-and-shuttle mechanism (8). This yeast Ncr1tunnel model also further supports previous work indicating thatmammalian NPC1 interacts with other sterol-shuttling proteins,including Gram1b on the ER membrane and ORP5 on the plasmamembrane, and that contact sites may be necessary for lipid exportfrom the lysosome (9, 10). Therefore, while these new models shedlight on how lipids might physically move from the vacuole, themechanisms and proteins involved in both the lipid traffickingdefect and accumulation in NPC remain unknown.

1Department of Pharmacology, University of Oxford, Oxford, UK 2Department of Molecular Genetics, Weizmann Institute of Science, Rehovot, Israel 3NuffieldDepartment of Obstetrics and Gynecology, University of Oxford, Oxford, UK 4Institute of Ophthalmology–Cell Biology, University College London, London, UK 5EuniceKennedy Shriver National Institute of Child Health and Human Development, National Institute of Health, Bethesda, MD, USA

Correspondence: frances.platt@pharm.ox.ac.uk

© 2020 Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 1 of 12

on 9 September, 2021life-science-alliance.org Downloaded from http://doi.org/10.26508/lsa.201800253Published Online: 2 June, 2020 | Supp Info:

In this study, we exploited the power of yeast genetics andperformed three independent systematic screens. Our objectiveswere to identify proteins that are affected by loss of Ncr1 andmaybecontribute to the pathology. This could be either through a physicalinteraction with Ncr1, by being indirectly affected at the level ofintracellular location, or by becoming essential for cellular phys-iology in the absence of Ncr1. Some of the genes identified in ourscreens are associated with cellular phenotypes reported previ-ously in NPC disease. These include calcium dysregulation, mito-chondrial dysfunction, metal ion homeostasis defects, and lipidtrafficking abnormalities. However, we also identified genes in-volved with the cytoskeleton and nutrient sensing, biologicalprocesses not previously linked to this disorder. We found thatcytoskeletal defects predicted by the yeast data occur in patient-derived cells, demonstrating the usefulness of yeast studies tofurther our understanding of NPC disease.

Results

Identification of Ncr1 interaction partners on the vacuolemembrane

To shed light on the pathology of NPC using yeast as a modelorganism, we performed three independent, unbiased screens(Tables S1–S3). The first screen focused on uncovering additionalinteracting proteins for Ncr1.

NPC1 is thought to transiently interact with NPC2 to exchangecholesterol via the N-terminal cholesterol-binding loop of NPC1 inthe lysosomal lumen (2). However, other interacting proteins(transient and more stable interactors) remain uncharacterized.We, therefore, performed a protein complementation assay toidentify proteins that physically interact with Ncr1 on the vacuolemembrane. We screened by the split-dihydrofolate reductase(DHFR) assay (11). Specifically, Ncr1 was tagged with one half of theDHFR enzyme and mated to strains carrying fusion proteins to theother half of the enzyme. Only if proteins physically interact will thisenable the complementation of the full DHFR enzyme and resis-tance to methotrexate, which inhibits the endogenous, essential,DHFR but not the synthetically encoded one. Because the DHFRfragments in this library are fused at the carboxyl terminus (C9), weassayed only the 48 proteins that are known to localize correctly tothe vacuole membrane when tagged at their C9 (12) (Table S1).Because only vacuolar membrane proteins were used in the screen,known interactors such as Npc2, which resides in the vacuole lu-men, were not observed. Of the 48 vacuolar membrane proteinsassayed, only three interacted significantly: Pmc1—a calciumATPase, Apc11—an anaphase-promoting complex member involvedin cell cycle regulation, and Fth1—an iron transport protein. Thestrength of their interactions with Ncr1 was measured as a functionof colony size divided by their abundance and was calculated to be0.4, 0.2, and 0.15 for Pmc1, Apc11, and Fth1 respectively, posing Pmc1as the most robust vacuolar membrane interactor of Ncr1 (Fig 1A).

To validate the interaction with Pmc1, we performed a reciprocalassay, using Pmc1 as the bait. This screen verified the interaction withNcr1, thereby confirming the physical association between the two

proteins. Strong interactions were again identified with the anaphase-promoting complex protein Apc11 and the iron transporter Fth1. Inaddition to these, we identified the ABC transporter Ycf1 as a stronginteracting partner with Pmc1 (Fig 1B). Based on these data of inter-action both between Pmc1 and Ncr1, as well as other shared proteininteraction partners Apc11 and Fth1, it suggests that these proteinsmaybe residing as a complex on the vacuolar membrane (Fig 1C). Theinteraction between Ncr1 and Pmc1 had also been previously sug-gested from a genome-wide protein–protein interaction screen (11);however, this interaction had not been validated.

Because our screens aimed to shed light on the mammalianNPC1 protein, we investigated the interaction of NPC1 with thehuman homologue of Pmc1, PMCA2/ATP2B2 in a mammalian sys-tem. We performed a co-immunoprecipitation from isolated ratcerebellum and demonstrated that PMCA2/ATP2B2 is pulled downtogether with NPC1, validating the yeast screen findings (Fig 1D).Furthermore, we examined Pmca2 (Atp2b2) transcript levels in theNPCm1n mouse model, which is null for the Npc1 protein because ofan insertion resulting in deletion of 11 of the 13 transmembranedomains resulting in a premature truncation of the Npc1 protein(13). A significant reduction in mRNA levels of Pmca2/Atp2b2 in theNpc1−/−mouse cerebellum as compared with the WT (P = 0.0004, Fig1E) was observed. This suggests that the loss of functional NPC1protein could affect the expression of proteins, such as PMCA2, thatit interacts with.

Pathways essential to sustain viability in Δncr1 yeastcorroborates a role for mitochondria in NPC

Changes in protein localization may result from primary or sec-ondary loss of cellular function and abnormal physiology in theabsence of Ncr1 or they may reflect a compensatory mechanism(s).For example, cholesterol accumulation and impaired transferrinreceptor trafficking in NPC1-deficient CHO cells can be corrected byoverexpression of acid sphingomyelinase, whose reduced activity isa secondary defect in these cells (14). To differentiate betweenthese options, we hypothesized that pathways that compensate forloss of Ncr1 would be essential for survival of the Δncr1 strain. We,therefore, performed a synthetic sick/lethal screen that compareda genome-wide yeast knockout library crossed onto the Δncr1background, with the same library crossed to a wild-type back-ground. This systematic synthetic sick/lethality screen identifiedmore than 50 proteins whose loss exacerbated the phenotype ofΔncr1, resulting in slower growth (Table S2).

Genes that were identified in this screen included those involvedin copper transport (MAC1), sphingolipid and fatty acid biosynthesis(FEN1 and HTD2), mitochondrial function (FZO1 and ILM1), andprotein sorting (MVB12) (hits found summarized in Table 1). Toconfirm these results, a serial dilution was performed on a Δncr1background with either MAC1 or FZO1 under a repressible promoter(GAL1pr) (Fig 2A). Indeed, growing the cells on glucose to represstranscription of MAC1 demonstrated a more severe growth phe-notype on the background of double mutants relative to the loss ofNcr1 alone. However, FZO1 expression is essential for normal yeastgrowth, so in fact, repressing FZO1 expression on the Δncr1 back-ground enhanced its growth, suggesting that the proteins operatein reciprocal pathways (Fig 2A).

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 2 of 12

Genes associated with mitochondrial function were identified inboth the mis-localization screen and the synthetic sick/lethalscreen. Interestingly, although an abundance of mitochondrialgenes were indicated, we did not observe any gross defects inmitochondrial morphology or function in the Δncr1 yeast by growingcolonies on glycerol, a non-fermentable carbon source (Fig S1A andB), suggesting any phenotypes may be quite subtle and compen-sated for by multiple mitochondrial pathways. In light of this, and

the expanding body of data demonstrating impaired mitochondrialhomeostasis in NPC disease (15, 16, 17, 18), we followed up ourfindings made in yeast by investigating relevant mitochondrialphenotypes in NPC1 patient cells.

One of the strongest hits identified in the synthetic sick/lethalyeast screen was ILM1, a gene essential for mitochondrial DNA(mtDNA) maintenance (19). We, therefore, quantifiedmtDNA in NPC1patient cells and observed a significant reduction (>20%) as

Figure 1. Identification of Ncr1 interaction partners on the vacuole membrane.(A) Ncr1 was tagged with one half of the DHFR enzyme and mated to strains carrying fusion proteins to the other half of the enzyme. Interaction strength was a functionof colony size of the diploids on methotrexate divided by their abundance. Ncr1 had high interaction strength with Pmc1 and Fth1 relative to the panel of vacuolarproteins. (B) Pmc1 was tagged with one half of the DHFR enzyme and mated to strains carrying fusion proteins to the other half of the enzyme and had high interactionstrengths with Fth11, Apc11, and Ncr1 relative to the panel of vacuolar proteins. Means ± SD, N = 8. (C) Graphic of predicted complex of Ncr1, Pmc1, Apc11, and Fth1. (D) Co-immunoprecipitation of NPC1 and PMCA2 (ATP2B2) from rat cerebellum. (E) qRT-PCR for mRNA expression of PMCA2 (ATP2B2) in wild-type, heterozygous, and homozygouscerebellums of 8-wk Npc1nih balb/c mice. N = 5, ***P = 0.0004 as compared with WT, calculated by one-way ANOVA.

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 3 of 12

compared with healthy controls (P < 0.001; Fig 2B). In addition, wefound hyperpolarization of the mitochondrial membrane potentialin NPC1 patient fibroblasts (P < 0.0001; Fig 2B) and an increase inperinuclear mitochondria in the NPC1 fibroblasts (P < 0.01; Fig 2B).Moreover, when we measured the rate of oxygen consumption inWT and NPC1-deficient CHO cells, we found that NPC1 null cellsgrown in 1 μM reduced glucose medium displayed significantlyreduced oxygen consumption as compared with controls (P <0.0001; Fig S1C).

Mitochondrial length is determined through a balance of fissionand fusion events (20). Interestingly, a gene identified in the syntheticsick/lethal screen, the GTPase FZO1 (mitofusin), is required for mi-tochondrial fusion (21, 22). Using high content-screening (23) to-gether with electron microscopy (Fig 2C), we determined meanmitochondrial length in NPC1 patient cells and confirmed that therewas a significant increase in length in comparison with controls(GM3123, 2.59 μm ± 0.077; NPC1+/+, 1.74 μm ± 0.043 **P < 0.01, n = 100;Fig 2D). Because the fusogen FZO1 is essential for growth in theabsence of Ncr1, it suggests that hyper fusion is a protectivemechanism that cells rely on in the absence of Ncr1/NPC. Fur-thermore, deletion of the fission gene (FIS1) in Δncr1 GAL1-FZO1double-mutant cells partially rescued the phenotype, suggestingthat it is the imbalance of mitochondrial fission and fusion that iscausing the lethality (Fig 2E).

Alterations in protein localization caused by loss of Ncr1 lead tocytoskeleton abnormalities in NPC

We hypothesized that the Δncr1 strain, which does not show asignificant growth phenotype, has modified or re-organized cel-lular networks to compensate for the loss of function of Ncr1. Usinga library in which each protein is tagged with GFP (12), we usedautomated mating approaches (24) to integrate the Δncr1 alleleinto the GFP library and, using a high-content microscopy setup(25) systematically compared the localization of all yeast proteins,when expressed on the Δncr1 background compared with controlcolonies.

We found thatmore than 40 proteins were localized to a differentorganelle in Δncr1 as compared with control strains (Table S3).Although somemay be a direct effect of Ncr1 loss, others may be theresult of adaptive changes to altered cellular physiology. Proteinsthat were found to have an altered organelle localization includedthose involved in copper sensing and regulation (Ctr1), vacuoleprotein sorting (HOPS and ESCRT: Vps41 and Did4), myosin motors(She4p/Dim1), mitochondrial respiration (Rsf1 and Rcf1), steroltransport (Pry1), peroxisome biogenesis (Pex17), nutrient sensing(Tco89), and actin cytoskeleton organization (She4 and Prk1) (cy-toskeletal proteins Prk1, She4 shown in Fig 3A, hits summarized inTable 1).

As we identified proteins associated with actin organization (Prk1and She4) to be mislocalized in the GFP screen, we hypothesizedthat as mitochondria move along actin filaments in budding yeast(26), cytoskeletal dysfunction could be contributing to the mito-chondrial dysfunction observed in the patient fibroblasts. In highereukaryotes, mitochondrial positioning is regulated by microtubules(27) (rather than actin in yeast), so we examined the microtubulenetwork in NPC-deficient mammalian cells.

One of the genes identified in the synthetic sick screen, ADA2, hasbeen shown in yeast to potentiate the acetyltransferase activity ofGcn5 (28). As mammalian GCN5 plays a role in α-tubulin acetylation(29) and mitochondria in mammalian cells move preferentially onacetylated microtubules (30), we first examined the acetylationlevel of the tubulin network in NPC1 patient fibroblasts compoundheterozygous for the most common human mutation I1061T, andalso R1186G. Patients with this mutation encode a functional NPC1protein, but it is then targeted for degradation due to proteinmisfolding (31). We examined the relative amount of acetylatedα-tubulin and observed a significantly higher level of acetylatedα-tubulin staining fluorescence in the NPC1 patient cells as com-pared with controls (representative images Fig 3B; P < 0.01 Fig 3C).

Western blotting confirmed the enhanced levels of acetylatedα-tubulin, together with greater amounts of α-tubulin acetyl-transferase (αTAT1) (Fig 3D and E). However, interestingly an in-crease in both the deacetylases, sirtuin 2 (SIRT2) and histone

Table 1. Specific proteins and genes extracted from yeast Ncr1 screens according to function or organelle association.

Function Screen for interactionpartners Mislocalisation screen Synthetic sick/lethal screen

Mitochondrial function Rcf1, Mcy1, Mrpl23, Aep1, Mss116, Nam2, Rsf1,and Coq3

SOM1, FMP37, MCT1, MRPL17, COA2, ATP20, ILM1,RSM22, STF2, HTD2, FZO1, ADK1, OCT1, and MSF1

Cytoskeleton She4 and Prk1 ADA2

Calcium homeostasis Pmc1 Pmr1

Metal ion homeostasis Fth1 Ctr1 MAC1

Lipid trafficking Pry1

Nutrient sensing and nutrientuptake Tco89 AGP1

Protein sorting Vps41, Did4, Pmr1, and Yip5 MVB12 and APS1

Peroxisome Pex17 AAT2

Lipid homeostasis Erg27 and Sel1 CHO2 and, FEN1

Cell cycle Apc11 WHI4

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 4 of 12

Figure 2. Pathways essential to sustain viability in ΔNcr1 yeast corroborates a role for mitochondria in NPC.(A) Serial dilution of yeast strains was performed on Δncr1 background with eitherMAC1 or FZO on GAL1pr repressible promoter and grown in glucose to inhibit proteinexpression. (B) The relative mitochondrial DNA (mtDNA) content, TMRM intensity, and perinuclear distribution of mitochondria were measured in control and NPC patientcells. Mean ± SD. N = 100 **P < 0.01, ***P < 0.001, ****P < 0.0001; t test. (C) Representative EM images of control (a) and NPC (b) mitochondria. Scale bar = 1 μm. (D) Statisticaldistribution of the length in sections of control NPC1+/+ (open columns) and NPC1−/− patient (filled columns) mitochondria acquired from analyses of EM data. ****P <0.0001, calculated by unpaired t test with Welch’s correction. (E) Serial dilution of Δncr1 on GAL1pr-FZO, FIS1 background in glucose to inhibit protein expression.

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 5 of 12

Figure 3. Alterations in protein localization caused by loss of Ncr1 lead to cytoskeleton abnormalities in NPC.(A) Representative images from GFP localization screen including Prk1-GFP and She4-GFP in control (BY4741) relative to Δncr1 yeast. Prk1 is localized to the actin corticalpatch in BY4741 yeast, but in the Δncr1 yeast Prk1 is found to be diffuse and mainly cytoplasmic. She4 is localized primarily to the cytoplasm where in the Δncr1 yeast, thelocalization is more punctate. (B) Representative images of NPC1 and control patient fibroblasts stained for acetylated α-tubulin. Scale bar: 10 μm. (C) Graphicalrepresentation of the fluorescent intensities of acetylated α-tubulin. Mean ± SD. N = 10–20 **P < 0.01; t test. (D) Protein levels of α-tubulin, acetylated α-tubulin, αTAT1,HDAC6, and SIRT2 were measured in control and NPC1 patient fibroblasts. (D, E) Quantitation of blots from (D) by densitometry. (F) Treatment with 125 μM NAD for 24 hreduced acetylated α-tubulin expression in NPC1 patient cells, fluorescence measured using a flow cytometer and analysed with FloJo software. Mean ± SD, N = 3 *P < 0.05;one-way ANOVA.

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 6 of 12

deacetylase 6 (HDAC6) (Fig 3D and E), was also observed, suggestingthat the cells may be attempting to compensate for the increase inacetylated α-tubulin in an attempt to restore homeostasis. Thissuggests that the hyperacetylation is a consequence of NPC1 lossbut is not favourable to the cell, highlighting this phenotype as apotential target for therapeutic intervention. We postulated thatfurther enhancement of SIRT2 activitymight reverse hyperacetylationof α-tubulin in NPC1 cells. We, therefore, treated patient cells withNAD, a co-factor of SIRT2, and observed normalization of the levels ofacetylated α-tubulin (Fig 3F), demonstrating this hyperacetylationcan be normalized.

NPC therapies significantly reduce levels of acetylated α-tubulinin disease fibroblasts

To determine whether therapeutic interventions could also reverse heacetylated microtubule phenotype in NPC disease cells, we treated NPC1patient fibroblasts with the European Medicines Agency-approved drugmiglustat, an inhibitor of glycosphingolipid biosynthesis that providesbenefit via substrate reduction (32, 33), and hydroxypropyl-β-cyclodextrin(HPβCD), a cholesterol-sequestering agent forwhich results fromaphaseI/II clinical trial inNPC1 have been reported (34). Indeed, hyperacetylationwas significantly reducedafter treatmentwithbothmiglustat andHPβCD,so that levels of acetylated a-tubulin were not significantly different fromcontrol cells (+50 μMmiglustat P > 0.9999, +250 μMHPβCD P = 0.3003) (Fig4AandB), supporting thehypothesis that theacetylation is adownstreameffect of losing NPC1 activity. Moreover, our results demonstrate cyto-skeletal defects from actin in yeast to microtubules in humans, sug-gesting cytoskeletal defects may be novel hallmarks of NPC1.

Discussion

We have investigated novel aspects of the cellular features of NPCdisease by taking advantage of the highly conserved yeast Ncr1protein orthologue of NPC1. NPC1 is an evolutionarily conservedprotein, with orthologues identified in both yeast and prokaryotes(13) that share homology with bacterial permeases of the resistance-nodulation-division family (35). Functionally, it has been establishedthrough the experimental demonstration that the yeast Ncr1 canrescue the cellular phenotypes of NPC1−/− patient fibroblasts (6) andthat the mammalian protein has pump activity when expressed in

Escherichia coli (36). A number of studies of Ncr1 have also extendedour understanding of the mammalian protein (6, 7, 37, 38) by ex-amining and detecting phenotypes in deletion strains.

To further the insights derived from this model organism, weperformed unbiased genetic screens in yeast with the confidencethat these assays were likely to yield data relevant to the biology ofmammalian NPC1 and elucidate which protein interactions wererelevant in contributing to the cellular phenotypes. We used threeseparate strategies to identify proteins that might have direct orindirect interactions with Ncr1: a complementation assay to detectproteins that physically interact with Ncr1 on the vacuolar mem-brane, a localization screen for proteins that are mislocalized whenexpressed on an Ncr1-deficient background, and a synthetic sick/lethal screen to distinguish proteins that may function in com-pensatory or parallel pathways and whose loss leads to impairedgrowth or lethality in Δncr1 yeast.

We identified three proteins (Pmc1, Apc11, and Fth1) that dis-played a robust interaction with Ncr1 on the vacuole membrane,more than 40 proteins that changed their localization pattern in Δncr1background andmore than 50 proteins whose presence is required fornormal growth in the absence of Ncr1 (Tables S1–S3). The genes andproteins implicated from the yeast screens fell into a number ofdiscrete categories: trafficking, nutrient sensing, calcium andmetal ionregulation, mitochondrial function, and cytoskeleton–the three lastones of which we followed up on.

We have previously reported that NPC is a disorder involvingaltered lysosomal calcium homeostasis (39), and this is due to theaccumulation of sphingosine that either inhibits lysosomal calciumuptake or promotes calcium leak. However, the identity of the protein(s)responsible for refilling the lysosome with calcium remains elusive.

The calcium ATPase Pmc1 was previously shown in a genome-widescreen for protein–protein interactions to interact with Ncr1 (11), theconfirmation of this interaction on the yeast vacuole is of particularinterest as its mammalian orthologues, ATP2B1-4, are members of afamily of plasma membrane ATPase (PMCA) calcium transporters (40).Interestingly, it has beendemonstrated that sphingomyelin accumulationimpairs PMCA activity, causing loss of calcium homeostasis, oxidativestress, and neurodegeneration (41). Storage of the same lipid occurs inNPC (39), as does loss of calcium homeostasis, oxidative stress, andneurodegeneration. These findings suggest that mutations in the NPC1protein may in turn have effects on the function of proteins that itphysically interacts with. Furthermore, although the identification of the

Figure 4. NPC therapies significantly reduce levels ofacetylated α-tubulin in disease fibroblasts.(A) Representative images of control and NPC1 patientfibroblasts treated with 50 μM miglustat or 250 μMHPβCD stained for acetylated α-tubulin. Scale bar: 10μm. (B) Quantification of the fluorescent intensities ofacetylated α-tubulin. Mean ± SD. N = 10–20 ****P <0.0001, ns P > 0.05 two-way ANOVA.

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 7 of 12

pumps and/or transporters that may fill the lysosome with calciumremains unknown, ATP2B2 is a potential candidate and its localizationshould be further studied.

In addition to the identification of proteins involved in calciumhomeostasis, the identification of multiple yeast proteins that areassociated with mitochondria was of particular interest. These findingsare in agreement with previous studies (7) that reported that loss ofNcr1 resulted in mitochondrial dysfunction as indicated by increasedoxidative stress, shorter lifespan, reduced oxygen consumption, andlower cytochrome c oxidase activity. We, therefore, further investigatedthese mitochondrial defects in patient-derived fibroblasts and con-firmed that NPC1 cells displayed lower mtDNA content, which may infact underlie the impaired oxygen consumption observed in CHO cells.Mutations in mtDNA are known to be associated with a wide range ofclinical mitochondrial diseases and a high percentage of these arelocated in mitochondrial tRNA genes (42). Intriguingly, the mitochon-drial tRNA protein Nam2wasmislocalized in Δncr1 yeast, andmutationsin the human orthologue (LARS2) are associated with conditions thatinvolve increased cell death and multisystem failure (43).

In addition, mitochondria from NPC1 patient fibroblasts hadincreased length and were hyperpolarized in comparison with controlfibroblasts. Although a previous report (18) suggested thatmitochondria inNPC cells were smaller than those of controls, no quantitative data sup-porting this conclusion was provided in that study. We measured mito-chondrial length using electron microscopy and observed a significantincrease in length. It has been proposed that increased mitochondrialfusion can help overcome low levels of stress (44), thus this increase inlength suggests that the loss of Ncr1 causes stress-induced mitochondrialfusion. This stress-induced mitochondrial fusion is a response in whichmitochondria raise cellular ATP in response to various insults by elongatingand hyperpolarizing (45). It has also been suggested that mitochondrialfusion is aprotectivemeasureagainstmtDNA loss/mutants, andenhancedfusion may be a compensation mechanism to the reduced mtDNA levels(46). Previous studieshave, however, observed fragmentedmitochondria inneurodegenerative diseases (47), so taken together with our data, thissuggests that it is most likely a disturbance to the fission/fusion dynamicbalance that results in mitochondria dysfunction.

These findings add to an expanding body of data showing loss ofmitochondrial homeostasis in NPC disease (15, 16, 48). Although wedo not have a complete understanding of the mechanistic linkbetween lysosome and mitochondrial dysfunction in NPC, there isintimate communication between the two organelles that enablesfunctional crosstalk (49, 50, 51). Data also suggest the possibility thatimpairment of lysosomal function may be further amplified byfeedback from dysfunctional mitochondria (52). Especially as in NPCdisease cells, there is an increase in lysosome–mitochondriamembrane contact sites (10), which may contribute to the increasedmitochondrial cholesterol (53) andmitochondrial dysfunction inNPC.

The novel identification in the screens of yeast genes associatedwith the cytoskeleton, such as PRK1, a serine/threonine kinase thatregulates the organization of actin (54) and SHE4, a regulator ofmyosin motor domains (55), suggest that impaired mitochondrialfunction in NPC could result from defective organelle transport. We,therefore, investigated whether there are cytoskeletal phenotypesin mammalian NPC cells, specifically the microtubule network as thisis key in regulating the movement of mitochondria in mammaliancells. Although in this study we focused on microtubules, other

alterations to the cytoskeleton in NPC mammalian cells have beendescribed. For example, vimentin, a phosphoprotein component ofintermediate filaments, is hypo phosphorylated in NPC, leading to itsdisorganization and disruption of vesicular transport (56).

The dynamic properties of microtubules are known to be af-fected by posttranslational modification of tubulin subunits (57),and we observed significantly increased levels of acetylatedα-tubulin in patient fibroblasts. The precise role of α-tubulinacetylation in the control of microtubule dynamics has not beenfully resolved (58), but increased acetylation is a hallmark of sta-bilized microtubules (59) and promotes flexibility (60), which in-creases organelle resistance to mechanical stress (61).

Although the acetyltransferase αTAT1 is responsible for acety-lation (62), the major regulators of deacetylation of α-tubulin atlysine 40 are histone deacetylase 6 (HDAC6) (63) and sirtuin 2(SIRT2) (64); loss of either leads to hyperacetylation (64, 65), whichwe observed. We were able to confirm increased levels of all threeregulators, consistent with gene expression data (38), which may beindicative of attempts to restore α-tubulin acetylation homeostasis.

The demonstration, here and by others (38), of increased ex-pression of HDAC6 in NPC cells is relevant because of the interest inthe use of HDAC inhibitors as treatments for NPC. Several studies (66)have shown that the drug vorinostat (suberoylanilide hydroxamicacid, SAHA, Zolinza) can normalize liver lipid homeostasis in NPCmodels. Vorinostat does not inhibit SIRT2, so would not affect thedeacetylase activity of the protein, but it does reduce HDAC6 ex-pression (67). In light of this, the relationship between microtubuleacetylation and organelle (both mitochondrial and lysosomal)dysfunction deserves further examination, as it is possible that HDACinhibitors could aggravate this particular NPC cellular phenotype.

In summary, we have exploited genetic methods in a simplemodel organism to identify proteins that interact—physically or genet-ically—with the yeast orthologueofNPC1, and that are likely tobe relevantto the functioning of themammalian protein. What is clear from the yeastdata is that Ncr1 likely acts as part of a protein complex with direct andindirect binding partners. This raises the possibility that some mutationsin theNPC1 proteinmaynot affect the primary function of this protein perse but prevent it from binding/interacting with other proteins that arerequired for the complex to functionasawhole. Thiswill bean interestingarea of future investigation and will shed light on how NPC1 and itsinteracting partners function in calcium and iron regulation, organelletrafficking, mitochondrial function, cytoskeleton organization, and nutri-ent sensing, as well as point to new potential therapeutic avenues for thetreatment of NPC.

Materials and Methods

Yeast strains and libraries

We performed three independent, unbiased screens detailed below.Genes thatwere identifiedare listed in Tables S1–S3. Knownor proposedactivities for the proteins they encode are listed according to theSaccharomyces Genome Database (https://www.yeastgenome.org).

All yeast strains in this study are based on the BY4741 laboratorystrain (68). Genedeletionwas performedusing the pFA6plasmid seriesand verified using PCR for the loss of the gene copy (69). GFP-tagged

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 8 of 12

strains were picked from the GFP library (12) and deletion strains werepicked from the yeast deletion library (70). Known or proposed ac-tivities for the proteins they encode are listed according to the Sac-charomyces Genome Database (https://www.yeastgenome.org).

Robotic library manipulations

All genetic manipulations on entire libraries were performed usingSynthetic Genetic Array (SGA) techniques (24, 71). To manipulatelibraries in 1536-colony high-density format, a RoToR bench topcolony arrayer (Singer Instruments) was used.

Yeast library screening

TheΔncr1 strain was constructed andmated against the yeast deletionlibrary (70) using SGA techniques that allow efficient introduction of atrait (mutation or marker) into systematic yeast libraries. SGA wasperformed as previously described (24, 71). Colony size was thenquantified using the Balony free software for the analysis of images ofplates containing arrays of yeast (the software package is maintainedby Barry Young at the University of British Columbia, Vancouver,Canada). For the GFP library screen, the colonies were moved to liquidmedium, and for high-throughput microscopic screening.

High-throughput microscopy

Microscopic screening was performed using an automatedmicroscopysystem (ScanR system; Olympus) as previously described (24). Briefly,images were acquired for GFP (excitation 490/20 nm; emission 535/50nm) and bright-field channels. After acquisition, the images weremanually reviewed in MATLAB vs.2012a 7.17 using compare2picturesV5script. As there were no co-localization markers, we assigned onlythose localizations that could be easily discriminated by eyes.

Protein complementation assay screen using the DHFR library

Strains were taken from the DHFR protein complementation libraries(11). In this library, Haploid strain “a” is Nat resistant (+Nat), whereashaploid strain “α” is hygromycinB (+Hygro) resistant. Pmc1 haploid strain“a” and “α” were used; however, because of no Ncr1 haploid strain “a,”only Ncr1-α was used. Haploid strains of either pmc1 or ncr1 werematedwith the vacuolar membrane protein library on YPD-rich media plates(n = 8). Diploid selection was done twice on plates containing selectionmarkers (+Nat, +Hygro). Diploids were then moved to metallux media(0.200 g Methotrexate + 20 ml DMSO +YPD) for 7 d to select for proteinsthat are physically interacting. Colony size was then quantified using theBalony free software for the analysis of images of plates containingarrays of yeast (the software package is maintained by Barry Young atthe University of British Columbia, Vancouver, Canada). Interactionstrength was calculated by dividing colony size by relative abundance.

Cells

Human NPC1-mutant fibroblasts were obtained from the National In-stitute of Health (NPC5; severity score 14, 1061T/R1186G) and from CoriellInstitute for Medical Research (GM22871 [1920delG/IVS9-1009G>A]).Control human dermal fibroblasts were acquired from Sigma-Aldrich

(Cat. no. 106-05A). Thefibroblastsweremaintained inDMEMwith 10%FCS,1%penicillin/streptomycin, and 1% L-glutamine. All cells were cultured at37°C with 5% CO2. Antibodies and reagents were sourced as follows:mouse anti-acetylated α-tubulin (6-11B-1; Santa Cruz Biotechnology),mouse monoclonal anti–α-tubulin (ab11304; Abcam). Quant-iTPicoGreen sdDNA (Thermo Fisher Scientific) tetramethylrhod-amine methyl ester (TMRM; Invitrogen).

Co-immunoprecipitation

Rat brain tissues were homogenized, in NP-40 cell lysis buffer (Invitrogen)supplemented with Halt Protease Inhibitor Cocktail (Thermo Fisher Sci-entific) using an electric homogenizer and passed through a 25-gaugeneedle 10 times, followed by constant agitation for 2 h at 4°C. The lysateswere cleared by centrifugation at 16,000g for 20 min at 4°C. 500 μl of eachsupernatant was incubated, overnight at 4°C, with SureBeads Protein Amagnetic beads (Bio-Rad) previously bound to 10 μg of NPC1 (NB400-148;Novus Bioscience) rabbit polyclonal antibody or 10 μg of IgG from rabbitserum (Sigma-Aldrich), following the manufacturer’s instructions. Thebeadswerewashedandeluted by incubation, at an indicated temperaturefor 10 min, with lysis buffer supplemented with Protein Loading Buffer.

Mitochondrial morphology quantification

Using the high-content IN Cell 1000 (500 cells acquired per well; GEHealthcare Life Sciences) analyzer, we quantitatively measuredfluorescent mitochondria labelling in control and NPC1 fibroblasts.Raw images were processed, and parameters were obtained using acustomized protocol in the IN Cell developer toolbox (GE Health-care Life Sciences) (23).

mtDNA analysis

DNAwas purified using the DNeasy blood and tissue kit (QIAGEN) andamplified on a Corbett real-time quantitative PCR machine usingTaqman universal PCR master mix. Sequences of the mitochondrialprimers (100 nM) were AGGACAAGAGAAATAAGGCC and TAAGAAGAG-GAATTGAACCTCTGACTGTAA and probe (6-FAM; 200 nM) TTCACAAAG-CGCCTTCCCCCGTAAATGA. Copy number was normalized to the levelsof the single copy nuclear gene amyloid β, using primers (300 nM)TTTTTGTGTGCTCTCCCAGGTCT and TGGTCACTGGTTGGTGGC and probe(200 nM; Yellow Yakima) CCCTGAACTGCAGATCACCAATGTGGTAG.

Cytoskeleton image acquisition

All image acquisition was completed with Leica TCS SP8 scanninglaser confocal microscope equipped with LAS X software. Z-stackimage series were projected for maximum intensity with Fiji-ImageJsoftware (National Institute of Health; version 1.46), and contrast/brightness for applied Look Up Tables for each channel were ap-plied from untreated, wild-type fibroblasts.

Measurement of oxygen consumption

Control and NPC1-deficient CHO cells (39) were maintained in completeDMEM-F12 growth medium and before analysis transferred to themedium containing reduced levels of glucose for 24 h. Oxygen

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 9 of 12

consumption was measured using the MitoXpress-Xtra assay (Luxcel)according to the manufacturer’s instructions. Readings were made at1min 30 s intervals for 13 h. Data were standardized against cell numberdetermined by propidium iodide staining.

Western blotting

Cell lysates and co-immunoprecipitated samples were separated on4–12% SDS–PAGE gels (Thermo Fisher Scientific) and then transferredto membrane using Bio-Rad Trans-Blot Turbo system. Membraneswere blocked; incubated with primary antibodies—acetylated α-tubulin (6-11B-1; Santa Cruz Biotechnology), ATAT1 (ab58742; Abcam),SIRT2 (12650; Cell Signalling Technology), HDAC6 (H-300; Santa CruzBiotechnology), PMCA2 (PA1-915; Thermo Fisher Scientific), and NPC1(NB400-148; Novus Bioscience); washed, incubated with appropriateHRP-conjugated secondary antibody and developed with chemilu-minescent substrate (Thermo Fisher Scientific). Images were capturedon a Bio-Rad ChemiDoc XRS+ system and quantified using ImageLabsoftware (Bio-Rad). Membranes were then stripped and re-probedwith primary antibodies.

Drug treatments and FACS analysis

Patient fibroblasts were treated with 50 μMmiglustat (Actelion) 72 h,250 μM hydroxypropyl-β-cyclodextrin (Sigma-Aldrich) 24 h and 125μM NAD for 24 h before analysis of levels of acetylated α-tubulin.Cells were fixed with 4% paraformaldehyde, permeabilized withFACS-perm buffer (BD Biosciences), stained with anti-acetylatedα-tubulin antibody, and analyzed on a BD FACS Canto II cytometerusing FACSDiva software. 10,000 cell events were collected, and themolecules of equivalent fluorescence (MEFL) were calculated using8-peak Rainbow calibration beads (559123; BD Biosciences).

Electron microscopy

Cells were processed according to Eden et al (2016) (72). In brief, thecells were fixed in 2% paraformaldehyde/2% glutaraldehyde for 30min, post-fixed in 1% osmium tetroxide and 1.5% potassium ferri-cyanide, and incubated in 1% uranyl acetate. The cells were thendehydrated and embedded in TAAB-812 resin. 70-nm sections wereviewed on a Jeol 1010 transmission electron microscope and imagescaptured with a Gatan Orius SC100B charge-coupled camera.

Supplementary Information

Supplementary Information is available at https://doi.org/10.26508/lsa.201800253.

Acknowledgements

This research received funding from the European Union Seventh Frame-work Programme (FP7 2007–2013) under grant agreement no 289278 -“Sphingonet” and Action Medical Research. FM Platt is a Royal SocietyWolfson Research Merit Award holder and a Wellcome Trust Investigator inScience. ME Fernandez-Suarez is a Royal Society Newton International

Fellow. N Platt is supported by the Wellcome Trust (202834/Z/16/Z). TheSchuldiner lab is also supported by a Volkswagen foundation “Life” grant. MSchuldiner is an incumbent of the Dr. Gilbert Omenn and Martha DarlingProfessional Chair in Molecular Genetics.

Author Contributions

A Colaco: formal analysis, investigation, methodology, and wri-ting—original draft, review, and editing.ME Fernandez-Suarez: investigation and methodology.D Shepherd: investigation and methodology.L Gal: investigation and methodology.C Bibi: investigation and methodology.S Chuartzman: investigation and methodology.A Diot: investigation and methodology.K Morten: methodology.E Eden: investigation and methodology.FD Porter: resources.J Poulton: supervision and methodology.N Platt: methodology and writing—review and editing.M Schuldiner: conceptualization, resources, supervision, investi-gation, and writing—review and editing.FM Platt: conceptualization, resources, supervision, funding ac-quisition, and writing—review and editing.

Conflict of Interest Statement

The authors declare that they have no conflict of interest.

References

1. Vanier MT (2010) Niemann-Pick disease type C. Orphanet J Rare Dis 5: 16.doi:10.1186/1750-1172-5-16

2. Infante RE, Wang ML, Radhakrishnan A, Kwon HJ, Brown MS, Goldstein JL(2008) NPC2 facilitates bidirectional transfer of cholesterol betweenNPC1 and lipid bilayers, a step in cholesterol egress from lysosomes.Proc Natl Acad Sci U S A 105: 15287–15292. doi:10.1073/pnas.0807328105

3. Platt FM (2014) Sphingolipid lysosomal storage disorders. Nature 510:68–75. doi:10.1038/nature13476

4. Scott C, Ioannou YA (2004) The NPC1 protein: Structure implies function.Biochim Biophys Acta 1685: 8–13. doi:10.1016/j.bbalip.2004.08.006

5. Carstea ED, Morris JA, Coleman KG, Loftus SK, Zhang D, Cummings C, Gu J,Rosenfeld MA, Pavan WJ, Krizman DB, et al (1997) Niemann-Pick C1disease gene: Homology to mediators of cholesterol homeostasis.Science 277: 228–231. doi:10.1126/science.277.5323.228

6. Malathi K, Higaki K, Tinkelenberg AH, Balderes DA, Almanzar-Paramio D,Wilcox LJ, Erdeniz N, Redican F, Padamsee M, Liu Y, et al (2004)Mutagenesis of the putative sterol-sensing domain of yeast NiemannPick C-related protein reveals a primordial role in subcellularsphingolipid distribution. J Cell Biol 164: 547–556. doi:10.1083/jcb.200310046

7. Vilaca R, Silva E, Nadais A, Teixeira V, Matmati N, Gaifem J, Hannun YA, SaMiranda MC, Costa V (2014) Sphingolipid signalling mediatesmitochondrial dysfunctions and reduced chronological lifespan in theyeast model of Niemann-Pick type C1. Mol Microbiol 91: 438–451.doi:10.1111/mmi.12470

8. Winkler MBL, Kidmose RT, Szomek M, Thaysen K, Rawson S, Muench SP,Wustner D, Pedersen BP (2019) Structural insight into eukaryotic sterol

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 10 of 12

transport through Niemann-Pick type C proteins. Cell 179: 1–13.doi:10.1016/j.cell.2019.08.038

9. Du X, Kumar J, Ferguson C, Schulz TA, Ong YS, Hong W, Prinz WA, PartonRG, Brown AJ, Yang H (2011) A role for oxysterol-binding protein-relatedprotein 5 in endosomal cholesterol trafficking. J Cell Biol 192: 121–135.doi:10.1083/jcb.201004142

10. Hoglinger D, Burgoyne T, Sanchez-Heras E, Hartwig P, Colaco A, Newton J,Futter CE, Spiegel S, Platt FM, Eden ER (2019) NPC1 regulates ER contactswith endocytic organelles to mediate cholesterol egress. Nat Commun10: 4276. doi:10.1038/s41467-019-12152-2

11. Tarassov K, Messier V, Landry CR, Radinovic S, Serna Molina MM, ShamesI, Malitskaya Y, Vogel J, Bussey H, Michnick SW (2008) An in vivo map ofthe yeast protein interactome. Science 320: 1465–1470. doi:10.1126/science.1153878

12. Huh WK, Falvo JV, Gerke LC, Carroll AS, Howson RW, Weissman JS, O’SheaEK (2003) Global analysis of protein localization in budding yeast.Nature425: 686–691. doi:10.1038/nature02026

13. Loftus SK, Morris JA, Carstea ED, Gu JZ, Cummings C, Brown A, Ellison J,Ohno K, Rosenfeld MA, Tagle DA, et al (1997) Murine model of Niemann-Pick C disease: Mutation in a cholesterol homeostasis gene. Science 277:232–235. doi:10.1126/science.277.5323.232

14. Devlin C, Pipalia NH, Liao X, Schuchman EH, Maxfield FR, Tabas I (2010)Improvement in lipid and protein trafficking in Niemann-Pick C1 cells bycorrection of a secondary enzyme defect. Traffic 11: 601–615. doi:10.1111/j.1600-0854.2010.01046.x

15. Kennedy BE, LeBlanc VG, Mailman TM, Fice D, Burton I, Karakach TK,Karten B (2013) Pre-symptomatic activation of antioxidant responsesand alterations in glucose and pyruvate metabolism in Niemann-PickType C1-deficient murine brain. PLoS One 8: e82685. doi:10.1371/journal.pone.0082685

16. Kennedy BE, Madreiter CT, Vishnu N, Malli R, Graier WF, Karten B (2014)Adaptations of energy metabolism associated with increased levels ofmitochondrial cholesterol in Niemann-Pick type C1-deficient cells. J BiolChem 289: 16278–16289. doi:10.1074/jbc.m114.559914

17. Wos M, Szczepanowska J, Pikula S, Tylki-Szymanska A, Zablocki K,Bandorowicz-Pikula J (2016) Mitochondrial dysfunction in fibroblastsderived from patients with Niemann-Pick type C disease. Arch BiochemBiophys 593: 50–59. doi:10.1016/j.abb.2016.02.012

18. Yu W, Ko M, Yanagisawa K, Michikawa M (2005) Neurodegeneration inheterozygous Niemann-Pick type C1 (NPC1) mouse: Implication ofheterozygous NPC1 mutations being a risk for tauopathy. J Biol Chem280: 27296–27302. doi:10.1074/jbc.m503922200

19. Kang CM, Jiang YW (2005) Genome-wide survey of non-essential genesrequired for slowed DNA synthesis-induced filamentous growth inyeast. Yeast 22: 79–90. doi:10.1002/yea.1195

20. Shaw JM, Nunnari J (2002) Mitochondrial dynamics and division inbudding yeast. Trends Cell Biol 12: 178–184. doi:10.1016/s0962-8924(01)02246-2

21. Rapaport D, Brunner M, Neupert W, Westermann B (1998) Fzo1p is amitochondrial outer membrane protein essential for the biogenesis offunctional mitochondria in Saccharomyces cerevisiae. J Biol Chem 273:20150–20155. doi:10.1074/jbc.273.32.20150

22. Hermann GJ, Shaw JM (1998) Mitochondrial dynamics in yeast. Annu RevCell Dev Biol 14: 265–303. doi:10.1146/annurev.cellbio.14.1.265

23. Diot A, Hinks-Roberts A, Lodge T, Liao C, Dombi E, Morten K, Brady S,Fratter C, Carver J, Muir R, et al (2015) A novel quantitative assay ofmitophagy: Combining high content fluorescence microscopy andmitochondrial DNA load to quantify mitophagy and identify novelpharmacological tools against pathogenic heteroplasmic mtDNA.Pharmacol Res 100: 24–35. doi:10.1016/j.phrs.2015.07.014

24. Cohen Y, Schuldiner M (2011) Advanced methods for high-throughputmicroscopy screening of genetically modified yeast libraries. MethodsMol Biol 781: 127–159. doi:10.1007/978-1-61779-276-2_8

25. Breker M, Gymrek M, Schuldiner M (2013) A novel single-cell screeningplatform reveals proteome plasticity during yeast stress responses. JCell Biol 200: 839–850. doi:10.1083/jcb.201301120

26. Fehrenbacher KL, Yang HC, Gay AC, Huckaba TM, Pon LA (2004) Live cellimaging of mitochondrial movement along actin cables in buddingyeast. Curr Biol 14: 1996–2004. doi:10.1016/j.cub.2004.11.004

27. Fu MM, Holzbaur EL (2014) Integrated regulation of motor-drivenorganelle transport by scaffolding proteins. Trends Cell Biol 24: 564–574.doi:10.1016/j.tcb.2014.05.002

28. Balasubramanian R, Pray-Grant MG, Selleck W, Grant PA, Tan S (2002)Role of the Ada2 and Ada3 transcriptional coactivators in histoneacetylation. J Biol Chem 277: 7989–7995. doi:10.1074/jbc.m110849200

29. Conacci-Sorrell M, Ngouenet C, Eisenman RN (2010) Myc-nick: Acytoplasmic cleavage product of Myc that promotes alpha-tubulinacetylation and cell differentiation. Cell 142: 480–493. doi:10.1016/j.cell.2010.06.037

30. Frederick RL, Shaw JM (2007) Moving mitochondria: Establishingdistribution of an essential organelle. Traffic 8: 1668–1675. doi:10.1111/j.1600-0854.2007.00644.x

31. Gelsthorpe ME, Baumann N, Millard E, Gale SE, Langmade SJ, Schaffer JE,Ory DS (2008) Niemann-Pick type C1 I1061T mutant encodes a functionalprotein that is selected for endoplasmic reticulum-associateddegradation due to protein misfolding. J Biol Chem 288: 8229–8236.doi:10.1074/jbc.m708735200

32. Platt FM, Neises GR, Dwek RA, Butters TD (1994) N-butyldeoxynojirimycinis a novel inhibitor of glycolipid biosynthesis. J Biol Chem 269: 8362–8365.

33. Lachmann RH, Platt FM (2001) Substrate reduction therapy forglycosphingolipid storage disorders. Expert Opin Investig Drugs 10:455–466. doi:10.1517/13543784.10.3.455

34. Ory DS, Ottinger EA, Farhat NY, King KA, Jiang X, Weissfeld L, Berry-Kravis E,Davidson CD, Bianconi S, Keener LA, et al (2017) Intrathecal 2-hydroxypropyl-beta-cyclodextrin decreases neurological disease progression in Niemann-Pick disease, type C1: A non-randomised, open-label, phase 1-2 trial. Lancet390: 1758–1768. doi:10.1016/s0140-6736(17)31465-4

35. Ioannou YA (2000) The structure and function of the Niemann-Pick C1protein. Mol Genet Metab 71: 175–181. doi:10.1006/mgme.2000.3061

36. Davies JP, Chen FW, Ioannou YA (2000) Transmembrane molecular pumpactivity of Niemann-Pick C1 protein. Science 290: 2295–2298. doi:10.1126/science.290.5500.2295

37. Berger AC, Hanson PK, Wylie Nichols J, Corbett AH (2005) A yeast modelsystem for functional analysis of the Niemann-Pick type C protein 1homolog, Ncr1p. Traffic 6: 907–917. doi:10.1111/j.1600-0854.2005.00327.x

38. Munkacsi AB, Chen FW, Brinkman MA, Higaki K, Gutierrez GD, Chaudhari J,Layer JV, Tong A, Bard M, Boone C, et al (2011) An “exacerbate-reverse”strategy in yeast identifies histone deacetylase inhibition as a correction forcholesterol and sphingolipid transport defects in human Niemann-Picktype C disease. J Biol Chem 286: 23842–23851. doi:10.1074/jbc.m111.227645

39. Lloyd-Evans E, Morgan AJ, He X, Smith DA, Elliot-Smith E, Sillence DJ,Churchill GC, Schuchman EH, Galione A, Platt FM (2008) Niemann-Pickdisease type C1 is a sphingosine storage disease that causes deregulationof lysosomal calcium. Nat Med 14: 1247–1255. doi:10.1038/nm.1876

40. Stafford N, Wilson C, Oceandy D, Neyses L, Cartwright EJ (2017) Theplasmamembrane calcium ATPases and their role as major new playersin human disease. Physiol Rev 97: 1089–1125. doi:10.1152/physrev.00028.2016

41. Perez-Canamas A, Benvegnu S, Rueda CB, Rabano A, Satrustegui J,Ledesma MD (2017) Sphingomyelin-induced inhibition of the plasmamembrane calcium ATPase causes neurodegeneration in type ANiemann-Pick disease.Mol Psychiatry 22: 711–723. doi:10.1038/mp.2016.148

42. Brandon MC, Lott MT, Nguyen KC, Spolim S, Navathe SB, Baldi P, WallaceDC (2005) MITOMAP: A human mitochondrial genome database–2004update. Nucleic Acids Res 33: D611–D613. doi:10.1093/nar/gki079

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 11 of 12

43. Pierce SB, Gersak K, Michaelson-Cohen R, Walsh T, Lee MK, Malach D,Klevit RE, King MC, Levy-Lahad E (2013) Mutations in LARS2, encodingmitochondrial leucyl-tRNA synthetase, lead to premature ovarianfailure and hearing loss in Perrault syndrome. Am J Hum Genet 92:614–620. doi:10.1016/j.ajhg.2013.03.007

44. van der Bliek AM, Shen Q, Kawajiri S (2013) Mechanisms of mitochondrialfission and fusion. Cold Spring Harb Perspect Biol 5: a011072. doi:10.1101/cshperspect.a011072

45. Tondera D, Grandemange S, Jourdain A, Karbowski M, Mattenberger Y,Herzig S, Da Cruz S, Clerc P, Raschke I, Merkwirth C, et al (2009) SLP-2 isrequired for stress-induced mitochondrial hyperfusion. EMBO J 28:1589–1600. doi:10.1038/emboj.2009.89

46. Chen H, Vermulst M, Wang YE, Chomyn A, Prolla TA, McCaffery JM, Chan DC(2010) Mitochondrial fusion is required for mtDNA stability in skeletalmuscle and tolerance of mtDNA mutations. Cell 141: 280–289.doi:10.1016/j.cell.2010.02.026

47. Knott AB, Perkins G, Schwarzenbacher R, Bossy-Wetzel E (2008)Mitochondrial fragmentation in neurodegeneration. Nat Rev Neurosci 9:505–518. doi:10.1038/nrn2417

48. Torres S, Balboa E, Zanlungo S, Enrich C, Garcia-Ruiz C, Fernandez-ChecaJC (2017) Lysosomal and mitochondrial liaisons in Niemann-Pickdisease. Front Physiol 8: 982. doi:10.3389/fphys.2017.00982

49. Demers-Lamarche J, Guillebaud G, Tlili M, Todkar K, Belanger N, GrondinM, Nguyen AP, Michel J, Germain M (2016) Loss of mitochondrial functionimpairs lysosomes. J Biol Chem 291: 10263–10276. doi:10.1074/jbc.m115.695825

50. Soto-Heredero G, Baixauli F, Mittelbrunn M (2017) Interorganellecommunication between mitochondria and the endolysosomal system.Front Cell Dev Biol 5: 95. doi:10.3389/fcell.2017.00095

51. Elbaz-Alon Y, Rosenfeld-Gur E, Shinder V, Futerman A, Geiger T,Schuldiner M (2014) A dynamic interface between vacuoles andmitochondria in yeast. Dev Cell 30: 95–102. doi:10.1016/j.devcel.2014.06.007

52. Wong YC, Ysselstein D, Krainc D (2018) Mitochondria-lysosome contactsregulate mitochondrial fission via RAB7 GTP hydrolysis. Nature 554:382–386. doi:10.1038/nature25486

53. Charman M, Kennedy BE, Osborne N, Karten B (2010) MLN64 mediatesegress of cholesterol from endosomes to mitochondria in the absenceof functional Niemann-Pick type C1 protein. J Lipid Res 51: 1023–1034.doi:10.1194/jlr.m002345

54. Zeng G, Cai M (1999) Regulation of the actin cytoskeleton organization inyeast by a novel serine/threonine kinase Prk1p. J Cell Biol 144: 71–82.doi:10.1083/jcb.144.1.71

55. Toi H, Fujimura-Kamada K, Irie K, Takai Y, Todo S, Tanaka K (2003) She4p/Dim1p interacts with the motor domain of unconventional myosins inthe budding yeast, Saccharomyces cerevisiae. Mol Biol Cell 14:2237–2249. doi:10.1091/mbc.e02-09-0616

56. Walter M, Chen FW, Tamari F, Wang R, Ioannou YA (2009) Endosomal lipidaccumulation in NPC1 leads to inhibition of PKC, hypophosphorylationof vimentin and Rab9 entrapment. Biol Cell 101: 141–152. doi:10.1042/bc20070171

57. Westermann S, Weber K (2003) Post-translational modificationsregulate microtubule function. Nat Rev Mol Cell Biol 4: 938–947.doi:10.1038/nrm1260

58. Janke C (2014) The tubulin code: Molecular components, readoutmechanisms, and functions. J Cell Biol 206: 461–472. doi:10.1083/jcb.201406055

59. Webster DR, Borisy GG (1989) Microtubules are acetylated in domainsthat turn over slowly. J Cell Sci 92: 57–65.

60. Xu Z, Schaedel L, Portran D, Aguilar A, Gaillard J, Marinkovich MP, Thery M,Nachury MV (2017) Microtubules acquire resistance from mechanicalbreakage through intralumenal acetylation. Science 356: 328–332.doi:10.1126/science.aai8764

61. Portran D, Schaedel L, Xu Z, Thery M, Nachury MV (2017) Tubulinacetylation protects long-lived microtubules against mechanicalageing. Nat Cell Biol 19: 391–398. doi:10.1038/ncb3481

62. Kalebic N, Sorrentino S, Perlas E, Bolasco G, Martinez C, Heppenstall PA(2013) alphaTAT1 is the major alpha-tubulin acetyltransferase in mice.Nat Commun 4: 1962. doi:10.1038/ncomms2962

63. Hubbert C, Guardiola A, Shao R, Kawaguchi Y, Ito A, Nixon A, Yoshida M,Wang XF, Yao TP (2002) HDAC6 is a microtubule-associated deacetylase.Nature 417: 455–458. doi:10.1038/417455a

64. North BJ, Marshall BL, Borra MT, Denu JM, Verdin E (2003) The human Sir2ortholog, SIRT2, is an NAD+-dependent tubulin deacetylase. Mol Cell 11:437–444. doi:10.1016/s1097-2765(03)00038-8

65. Zilberman Y, Ballestrem C, Carramusa L, Mazitschek R, Khochbin S,Bershadsky A (2009) Regulation of microtubule dynamics by inhibitionof the tubulin deacetylase HDAC6. J Cell Sci 122: 3531–3541. doi:10.1242/jcs.046813

66. Munkacsi AB, Hammond N, Schneider RT, Senanayake DS, Higaki K,Lagutin K, Bloor SJ, Ory DS, Maue RA, Chen FW, et al (2017) Normalizationof hepatic homeostasis in the Npc1(nmf164) mouse model of Niemann-Pick type C disease treated with the histone deacetylase inhibitorvorinostat. J Biol Chem 292: 4395–4410. doi:10.1074/jbc.m116.770578

67. Kazantsev AG, Thompson LM (2008) Therapeutic application of histonedeacetylase inhibitors for central nervous system disorders. Nat RevDrug Discov 7: 854–868. doi:10.1038/nrd2681

68. Brachmann CB, Davies A, Cost GJ, Caputo E, Li J, Hieter P, Boeke JD (1998)Designer deletion strains derived from Saccharomyces cerevisiaeS288C: A useful set of strains and plasmids for PCR-mediated genedisruption and other applications. Yeast 14: 115–132. doi:10.1002/(SICI)1097-0061(19980130)14:2<115::AID-YEA204>3.0.CO;2-2

69. Longtine MS, McKenzie A 3rd, Demarini DJ, Shah NG, Wach A, Brachat A,Philippsen P, Pringle JR (1998) Additional modules for versatile andeconomical PCR-based gene deletion and modification inSaccharomyces cerevisiae. Yeast 14: 953–961. doi:10.1002/(sici)1097-0061(199807)14:10<953::aid-yea293>3.0.co;2-u

70. Giaever G, Chu AM, Ni L, Connelly C, Riles L, Veronneau S, Dow S, Lucau-Danila A, Anderson K, Andre B, et al (2002) Functional profiling of theSaccharomyces cerevisiae genome. Nature 418: 387–391. doi:10.1038/nature00935

71. Tong AHY, Boone C (2006) Synthetic genetic array analysis inSaccharomyces cerevisiae. Methods Mol Biol 313: 171–192. doi:10.1385/1-59259-958-3:171

72. Eden ER, Sanchez-Heras E, Tsapara A, Sobota A, Levine TP, Futter CE(2016) Annexin A1 tethers membrane contact sites that mediate ER toendosome cholesterol transport. Dev Cell 37: 473–483. doi:10.1016/j.devcel.2016.05.005

License: This article is available under a CreativeCommons License (Attribution 4.0 International, asdescribed at https://creativecommons.org/licenses/by/4.0/).

Yeast as a model for NPC disease Colaco et al. https://doi.org/10.26508/lsa.201800253 vol 3 | no 7 | e201800253 12 of 12